Extrahepatie versus intrahepatic cholestasis: is it possible to differentiate?

ABSTRACT

Accurate and prompt differentiation of extrahepatic cholestasis [extrahepatic biliary atresia [EHBA] and choledochal cyst] from other causes of intrahepatic infantile cholestasis is critical in determining the therapeutic outcome of these babies as in the former case surgical correction must be undergone as soon as possible. The aim of this work is to evaluate the various modalities used for making this differentiation by finding the sensitivity, specificity and accuracy of each. This was done through retrospective analysis of data [that included the history, the examination and the investigations] collected from the files of hundred consecutive infants whose ages ranged between one day and 6 months and who presented to our hepatology unit [a tertiary referral unit]. It was found that EHBA was diagnosed in 47% of infants and choledocal cyst in 3% both of which constituted the surgical causes [group I]. On the other hand medical causes [group II] accounted for 50% of the cases [idiopathic neonatal hepatitis in 27% and other causes as inborn errors of metabolism, infections, paucity of intrahepatic bile ducts and low GGT cholestasis in 23%]. The persistent presence of clay stools was statistically more common in group I compared to group II with a sensitivity of 94%, specificity of 52%, 73% accuracy and a 66% positive predictive value, while pruritus was significantly more common in group II compared to group I. Although the presence of hepatomegaly was not statistically different between the two groups yet, the presence of firm hepatomegaly was significantly more commonly encountered in patients of group I compared to those of group II. The only laboratory investigations that showed a statistical difference between the two groups were GGT and prothrombin time as both were significantly higher in group I babies while ALT, AST, serum bilirubin and serum albumin showed no statistical difference between the two groups. HIDA scan of the liver revealed a poor uptake among group II babies while group I babies showed no excretion of the dye in the intestine with a 100% sensitivity, 55% specificity, 78% accuracy and a 100% negative predictive value. Abdominal U/S revealed that the presence of the triangular cord sign [TC sign] was more common in group I as compared to group II with 80% sensitivity, 85.7% specificity, 83.3% accuracy and 85% negative predictive value. As regards the liver biopsy it was found that a disturbed architecture and bile ductular proliferation were significantly more commonly present in group I babies compared to group II ones with a sensitivity of 100%, specificity of 80%, accuracy of 91% and a 100% negative predictive value. It is thus concluded that standard tests of liver function are not discriminating between surgical and medical causes of cholestasis. Liver biopsy has the greatest diagnostic accuracy however no single test appears to be clearly superior in the differentiation of conditions leading to cholestasis; the use of several tests may improve the diagnostic power